Implementation of evidence-based practices (EBPs) within juvenile justice (JJ) settings has been limited. Service delivery occurs within disconnected practice settings (e.g., parole/probation, drug treatment, public health), with limited continuity of care. Effective strategies are needed for implementing EBPs to improve service delivery to adolescents under juvenile justice supervision and promote sustainable changes across large, multifaceted systems. The goal of the TRIALS initiative is to support implementation research that improves the continuum of substance abuse and HIV/AIDS services for adolescents in the legal system. Our research concept focuses on improving HIV/AIDS testing and referral within the JJ continuum. Using a randomized cluster design, the effectiveness of Community-wide (CW) and Horizontal Cascading (HC) implementation strategies will be compared. Both involve a Plan-Do-Study-Act approach that promotes change at the community level through training, the use of Community Change Teams (CCT), and guidance from external coaches. In the CW condition, JJ and other professionals (actively involved in overseeing and/or providing services for juveniles) will be asked to collaborate with multiple substance use treatment provider agencies within a specified community in the development of strategic plans to address gaps in the service delivery continuum. These individuals will be invited to participate in CCT training and planning efforts. Implementation of strategic plans will be allowed to proceed without additional coaching or implementation intervention strategies. In the HC condition, along with JJ and other professionals actively involved in providing services to juveniles, CCT membership will include a single treatment provider agency. The CCT will be tasked with developing a ?prototype? for carrying out the strategic plan within the participating treatment provider agency. Once implemented successfully, the prototype will serve as a model for systematic roll out to the other providers in the community, with CCT members functioning as ?in-vivo? coaches. The research concept involves random assignment of 40 clusters of providers (e.g., Community/county) to either the CW or HC study condition. Using mixed-methods, data will be collected at the community, provider, staff, and client levels. The degree to which implementation strategies, organizational readiness for change, and interagency collaboration promote implementation, service, and client outcomes will be examined using multilevel analytic techniques. In collaboration with the Texas Juvenile Justice Department, we have established an interdisciplinary team to design meaningful studies that will identify implementation strategies that are viewed as valuable, acceptable, feasible, and sustainable by the field.